Scientists at the Institute of Advanced Study in Science and Technology (IASST), an autonomous institute of the Department of Science and Technology (DST), have planned to develop a model that would be practically applicable in real-world situation and have unmatched accuracy in minimum amount of computation time.
Dr. Lipi B. Mahanta and her team experimented with various color models, transformation techniques, feature representation schemes, and classification methods to develop a powerful machine learning (ML) framework. The aim of this extensive analysis and experimentation was to identify the optimal combination for detecting cervical dysplasia.
The performance of the model was tested on two datasets, the first one collected from healthcare centres in India and the second one was a publicly available dataset. Using a method of image processing – the Non-Subsampled Contour Transform (NSCT) and the YCbCr colour model (a way of representing colours in an image), the new model achieved an average accuracy of 98.02%.
The findings published by MDPI in the journal 'Mathematics' highlighted the potential of their computational model to revolutionise the detection of cervical dysplasia. The innovative model could revolutionise the detection of cervical dysplasia and provide healthcare experts with highly accurate tools for improved diagnosis precision and better treatment outcomes.